**1. Introduction**

Predictive power of the yield curve is certainly not a new topic in finance. It has been empirically shown that a positive slope of the yield curve is linked with a future increase in real economic activity in the United States of America (US) in the early works by Estrella and Hardouvelis [1] while Estrella and Mishkin [2, 3] showed that yield curve provides a relatively strong signal in predicting US recession, particularly the one that had occurred during 1990–1991. In addition to that, Karunaratne [4] also finds that yield curve is a better and near perfect tool for forecasting economic activities compared to other macroeconomic indicators in Australia, whereas Hamilton and Kim [5] reexamined the predictability of the yield curve with updated data and confirmed the usefulness of the yield curve in predicting gross domestic product (GDP) growth. Apart from the above, there are voluminous studies that have been covering on the predictive ability of the yield curve. A much recent study by Boukhatem and Sekouhi [6], for example, also highlights that the yield curve can be considered as an advanced indicator for growth or recession for the Tunisian economy. As such, the slope of the yield curve, which typically referred to the yield spread or term spread, is deemed to act as a valuable forecasting tool<sup>1</sup> .

curve. Section 3 discusses the major crises faced by the Malaysian economy, within our data span while Section 4 describes the estimation model, data and method employed. The empirical results are presented in Section 5 while the concluding remarks and recommendation for future research are presented in Section 6.

*Has the Yield Curve Accurately Predicted the Malaysian Economy in the Previous Two Decades?*

According to the economic research department of the Federal Reserve Bank of New York, the analysis on the behavior of interest rates of different maturities over the business cycle goes back to the early work of Mitchell [13]. Many years later, Butler [14] made a connection between the yield curve as a predictor of short-term interest rates and the implications of declining short-term rates for contemporaneous economic activity, of which he correctly predicted that there would be no

Fundamentally, yield curve analysis is developed based on the term structure of interest rates, strongly associated with pure expectation theory. This theory essentially equalizes the long-term interest rates with short-term interest rates and market expectation of future interest rates plus a risk premium, which refers to the opportunity cost and compensation for holding long-term bonds as investors generally prefer short-term rather than long-term bonds. The linkage between the long-

predictability of the yield curve in charting the future state of the economy.

term and short-term rate together with the risk (or liquidity) premium is as

*int* <sup>¼</sup> *it* <sup>þ</sup> *iet*þ<sup>1</sup> <sup>þ</sup> *iet*þ<sup>2</sup> <sup>þ</sup> … <sup>þ</sup> *iet*þð Þ *<sup>n</sup>*�<sup>1</sup> *n*

where *int* is the long-term rate, it is the current short-term rate; *iet+(n-1)* is the future short-term rate; and *lnt* is the risk (liquidity) premium, which posits that the yield on a long-term bond is the average of the one period interest rates expected over the lifetime of the long bond. Hence, this theory puts forward that the expectations of market participants are to be formed rationally, based on the anticipated economic situation, leading to the expected level of future short rates that would in

Therefore, the link between the yield curve and growth of the economy is rationalized through the monetary policy actions undertaken by the government. For example, suppose the government undertakes a contractionary monetary policy. During this time, financial market participants would expect the short-term interest rates to be temporarily raised. If the current short-term interest rate is higher than the expected future short-term rate, the long-term rate should rise less than the short-term rate according to the expectations theory. Thus, the yield spread will be flattened, or, in extreme cases, be negative. This will also reflect situations where the yield curve will be inverted as the short-term rate is higher than the long-term rate. Inverted (or negatively sloped) yield curves have been excellent predictors of recessions for many years, whereby every recession after the mid-1960s was predicted by a negative slope—an inverted yield curve—within six quarters of the impending recession. During those times, the monetary contraction would eventually reduce spending in interest-sensitive sectors of the economy,

<sup>4</sup> Incorporating the modification into the expectation theory, with the risk/liquidity premium, widely

. Subsequently, numerous studies have been conducted on the

þ *lnt* (1)

**2. Theoretical framework of yield curve analysis**

*DOI: http://dx.doi.org/10.5772/intechopen.92214*

recession in 1979<sup>3</sup>

presented in the equation<sup>4</sup> below;

turn influence the yields on long-term bonds.

<sup>3</sup> Federal Reserve Bank of New York.

**153**

known as liquidity premium theory, see Mishkin and Eakins [11].

Even though the use of the yield curve in predicting the economy has been empirically proven, recent research came up with new evidence questioning whether it is still as powerful. Chinn and Kucko [7], for example, show the predictive power of the yield curve in the United States of America (US) and Japan to be declining over time. In addition to that, they also show that for all of the seven countries examined, yield spread is indeed important and has significant predictive power when forecasting industrial production growth over a 1-year horizon, but the result deteriorates when forecasting growth 2 years ahead.

In Malaysia, the Treasury bill spread have been empirically shown to be a significant predictor of future growth of annual output, see Ghazali and Low [8], while Elshareif and Tan [9] find a long-run cointegrated relationship between the short- and long-term rates, confirming the existence of pure expectation theory in the Malaysian bond market. The recent work of Zulkhibri and Abdul Rani [10] is one of the first to examine the role of yield spread<sup>2</sup> in inflation and growth in Malaysia. Based on the data span of 1992–2009, they used simple regression to establish the relationship between yield spread, output and inflation, and then used the probit model for forecasting. It was shown that the yield spread contains little information about future output and inflation at short horizons. They also argued that the use of yield spread in monetary analysis beyond conventional indicators is rather limited.

This study aims to discover the ability of the yield spread to predict economic growth over a longer time horizon. In consideration that the Malaysian economy as well as the sovereign bond market has grown rapidly over the past 20 years, it would be interesting to see whether there exists a long-run relationship between the yield spread and growth.

In addition to that, this study also aims to use the autoregressive distributed lag (ARDL) approach to cointegration and error correction models (ECMs) to determine whether there is evidence of relationship between yield spread and growth, in long run and short run within the span of the 20 years. In consideration of the prerequisites possessed by the conventional Granger [11] and, Engle and Granger [12] to have all the underlying variables to be in the same order of integration, ARDL stands out to be the most appropriate technique in order to test for cointegration among the variables.

This paper makes two contributions to the existing literature. First, it examines the relationship between slope of the yield curve (in other words, yield spread) and growth based on updated data and over a 20-year time period. Second, it is the first to employ the ARDL method in consideration of the mixture of order of integration among the variables tested. The empirical result shows the existence of a long-run relationship between the yield spread and growth in Malaysia. Though significant, the instability of the yield spread to affect the movement of growth does not support the priori expectation on the predictive power of the yield curve, making it less reliable to be used as forecasting tool on the general economic condition.

The remainder of this paper is organized as follows. Section 2 highlights the theoretical framework and related literature on the predictive power of the yield

<sup>1</sup> Normally calculated as the difference between the yields of the 10-year government securities against 3-month Treasury bill. See among others, Estrella and Hardouvelis [1], Estrella and Mishkin [2], Karunaratne [4], Hamilton and Kim [5].

<sup>2</sup> Zulkhibri and Abdul Rani [10] used 'term spread' in their study instead of yield spread, but the calculation is similar as highlighted earlier.

curve. Section 3 discusses the major crises faced by the Malaysian economy, within our data span while Section 4 describes the estimation model, data and method employed. The empirical results are presented in Section 5 while the concluding remarks and recommendation for future research are presented in Section 6.
